Stories that deal with airflow always start out with the classic definition that an engine is really just a simple air pump—the air goes in and it comes out. This chapter is no different because that’s the most simplistic way to approach an internal combustion engine. In order to do this, we should first know some fundamental facts about the working fluid we will be dealing with.
This Tech Tip is From the Full Book, HIGH PERFORMANCE CHEVY SMALL-BLOCK CYLINDER HEADS. For a comprehensive guide on this entire subject you can visit this link:
SHARE THIS ARTICLE: Please feel free to share this article on Facebook, in Forums, or with any Clubs you participate in. You can copy and paste this link to share: http://www.chevydiy.com/airflow-basics-for-chevy-small-block-cylinder-heads/
The atmosphere we breathe and that our performance small-block Chevy inhales is comprised of roughly 21 percent oxygen, 78 percent nitrogen, and one percent miscellaneous components. The part we are most concerned with is the oxygen, since this is the part that actively contributes to combustion. The more air we can stuff into the cylinders and keep there, the more potential power our engine will make. But there’s more to this equation than just sheer volume. Atmospheric conditions play a part in this equation as well. The three critical ones are air temperature, air pressure, and the amount of water present in the air. This is generally referred to as humidity, but the engineers like to call it vapor pressure. A discussion of each of these conditions is important since all three play a part in what is generally referred to as air density.
The air that encompasses this planet we live on has both volume and weight. The weight of this air is directly related to where you measure it. For example, if we take one square inch area and measure the pressure at sea level, this column of air will weigh 14.7 pounds per square inch (psi). Up in the mountains, a similar column of air will weigh much less, perhaps only 12.5 psi, because we have shortened the standing height of the air. So the point here is that air pressure changes based on altitude relative to a sea-level standard. With less pressure, there is less atmospheric force pushing the air into our engine. This reduces the amount of air in the cylinders, producing less power. Temperature is an equally important factor in this trilogy. As the temperature of the air increases, the density of oxygen molecules in a cubic foot of air is reduced. This gives us less air to combust the fuel. So it’s safe to say that higher inlet air temperatures generally produce less power. The last component is the amount of water vapor present in the air going into the engine. It’s no secret that water doesn’t burn, so the volume of water present in the air not only displaces a certain amount of air, but also contributes to reducing combustion efficiency, which means less power.
This leads us to a concept called standard temperature and pressure (STP) that is most often used in the performance industry. STP is defined as 60 degrees of dry air (zero vapor pressure) at sea level, or 14.7 psi of air pressure. This combination of atmospheric conditions is used as the atmospheric standard for engine dyno work. No matter what the existing conditions are during a dyno test, the horsepower and torque will be “corrected” to that atmospheric standard. The correction factor is applied to the observed torque and horsepower numbers, which usually means that the corrected numbers are higher than the observed numbers, assuming the existing atmospheric conditions are worse than the STP. This standard used to be the correction factor for original equipment manufacturers (OEM) until the 1970s when more realistic temperature, pressure, and the inclusion of vapor pressure, reduced the corrected power levels slightly. There have been several revisions to this standard, but generally, if you subtract 5 percent from a STP-corrected data, you will be very close to what the OEM’s now use as their horsepower correction factor.
Now that we have applied a standard to the air that’s moving into the engine, we can now begin the investigation of how that air moves through the engine, and, for the subject of this book, through the cylinder heads of a small-block Chevy. In order to quantify size in the cylinder head market, port volume has become the most popular shorthand way to identify port size. While this continues to be the popular shortcut to identify a head, it is not the best. Generally speaking, larger port volumes tend to flow more air and are therefore considered a better application for higher engine speeds and more peak power. In reality, this is a gross overgeneralization and far too simplistic. Frankly, if selecting cylinder heads were that easy, there would be little call for the information in this book!
While port volume does work to categorize heads, and we use that throughout this book, it’s merely a signpost to direct the more knowledgeable cylinder head user to more specific data that he can use. We’ve included a couple of sidebars that deal specifically with intake port cross-sectional area and those two sidebars are among the most important pieces of information in this entire book (“Cross-Sectional Areas” and “Power Secrets”). If you take the time to evaluate any intake port not only by both its cross-sectional area and its port flow, that evaluation technique will always serve to point you in the right direction when it comes time to evaluate and select a cylinder head. Now that we have a basis from which to work, it’s time to get down to what really happens in both the intake and exhaust ports.
This section looks at the basics of airflow in the intake port. The goal here is to explain some basic flow examples, but in no way can this be construed as anything more than a very basic look at airflow. This is a complex and highly specialized area and certainly full of differing theories, many of which are guaranteed to change as we continue to learn more about how an internal combustion engine operates. We can start with the concept that much of the information delivered here has been determined with help from a flow bench, and therein rests the first dilemma. A flow bench, by design, is a steady-state flow measurement tool. In other words, it measures air movement primarily in one direction, either intake or exhaust, and does little to address the intermittent motion and major pressure excursions that are present in a running engine. As a result, many of our theories on airflow, while they may seem to make sense, are in fact flawed. That is what leads us down the path of discovery. This may sound like a giant disclaimer, and that is partially true. The point is that much of what we know is based on assumptions that the future may prove to be not entirely accurate. But it’s what is known as contemporary wisdom, and that’s the best we can do with what we know.
If we accept that velocity and mass flow are key components in our attempt to shove air and fuel into a cylinder, then it would seem logical to look at what happens to air as it attempts to move from the intake manifold on its way past the intake valve. In the intake port, we have both air and some ratio of fuel mixed with the air. We must also assume that this mixture of air and fuel will behave differently when required to change direction. Physics and common sense tell us that the heavier fuel droplets are less likely to make a change in direction than the lighter air molecules. This means that in order to create a more efficient intake port, we need to ensure that any change in direction includes more of the fuel droplets combined with the air. The more homogenous the mixture, the more efficiently (and quicker!) that mixture will burn in the cylinder.
So how do we accomplish this? Let’s start with a poor design and work our way toward a more efficient configuration. Let’s start with the current small-block Chevy architecture and what we’re forced to work with. These design parameters (some may call them restrictions) include the stock 23-degree valve angle, the intake port angle at the manifold, and limitations on port width because of pushrod placement. These are the realities of the layout of a small-block Chevy. The stock valve angle means that we must force the air and fuel to make a significant change in direction in order to transition past the intake valve. This is measured as the angular valve face relationship to the cylinder. Had the first generation small-block Chevy design engineers realized the value of a less acute angle, we might have been blessed with a taller intake valve angle such as the GEN III small-block, beginning in 1998, came with a more efficient 15-degree valve angle. Of course, airflow was not the only consideration when Chevy designed the small-block. We’ll save that discussion for some other time. To show how important this is, in a matter of a scant few years, GM changed the 15-degree valve angle on the LS2 to 12 degrees for the LS7 on its way to an amazing 505 SAE production horsepower.
So given this 23-degree valve angle, one of the first priorities for a port designer is to attempt to “fool” the air into thinking that it has made less of a change from straight ahead flow as possible. Much like auto racing, the wider the radius of the turn, the faster you can negotiate that corner. As we look at older small-block Chevy intake ports, one thing that is quickly identified is that those older ports tend to have very flat floors. This is not a big problem until it comes time for the air to make the change in direction into the bowl just above the valve seat area. Air (and fuel) traveling across this sharp ledge quickly separates, and moves across the valve head toward the roof of the port. Any change in direction of a port will witness this type of movement, but it is especially bad in flat floor ports.
Most port designers and porters refer to the floor of the port where it transitions into the throat area as the short turn radius. At low port velocities, a gentler short side radius helps to maintain contact with the air stream, and this is what often results in improved low and mid-lift flow performance. As velocity and mass flow increases, there comes a point where the air still separates from the short-side radius. The higher the inlet velocity where this separation occurs, the more dead area is created by this separation, which is a major condition in the port’s flow curve performance. Of course, increasing the height of the short-side radius also reduces the cross-sectional area of the port, which demands that the top of the port be raised to accommodate the taller short side. The problem here is that the spring seat and valveguide placement limit a taller port roof. This is also determined by the 23-degree valve angle. A 15 or 18-degree valve angle allows more room for this taller roof, which is another reason why a taller valve angle works better for highRPM engines.
As the air (and fuel) makes the transition into the valve bowl area, almost all intake ports increase the port volume in this area to slow the air to help it turn the corner. Still with sufficient velocity even at partial valve lift, air tends to be pushed toward the high side of the bowl area. This is an inevitable result of increased velocity through the port, which tends to concentrate a majority of the flow around the tall side of the intake valve. If you picture the intake valve at a given lift—say 0.500 inch for example—the circumference of the valve times the lift create a flow window sometimes called a flow curtain area. Adding a larger intake valve increases this flow window by a factor of pi (3.1417), which is one reason why adding a larger valve generally increases flow. In an ideal world, the flow distribution around the intake valve into the cylinder would be symmetrical, meaning flow and a pressure distribution around the valve would be the same at every point around the valve. As you can imagine, that rarely happens. One key to a more efficient cylinder head is an intake valve where more of the air and fuel moves past the short-side portion of the valve, and this is what the port designer seeks to achieve.
Tracking the movement of air and fuel in the intake port, we now find the air making that important move past the intake seat, moving between the seat angles and the angle on the intake valve. While every point in the intake port is important, it’s hard to find a single area that is more important than the area surrounding the interface between the valve and the seat. The classic approach is a three-angle seat that begins in the port throat area with a 60-degree angle. The sealing angle that contacts the valve is 45 degrees and then transitions to a top cut (closest to the combustion chamber) of anywhere from 35 to 15 degrees, depending upon the application. While we use three angles as an example, the cylinder head designer is certainly not limited to just three angles. Many performance valve jobs now incorporate up to five angles.
The idea with multiple seat angles is to maximize airflow through the valve curtain area throughout the entire lift curve. This is a critical distinction that deserves discussion because most enthusiasts tend to focus far too much attention on peak valve lift flow numbers, ignoring the low and mid-lift flow numbers. We emphasize this throughout the book, not to bore you but to reinforce just how critical these flow numbers are to overall performance. This is how the three-angle seat evolved. In fact, there is data to support that perhaps a 70 to 75-degree transitional throat cut underneath the 60-degree seat angle is beneficial in certain applications.
Another important point here is the transient nature of all these numbers. Given unlimited time on the flow bench, the goal would be to experiment one degree at a time with each seat angle, along with different valve seat and back-cut angles. For most enthusiasts, this is hardly practical. But we know of many drag racers and cylinder head experts who have performed this exercise. The frustrating part of this process is that as soon as the port is even slightly changed, the very specific valve seat angles do not perform the same way. Often, the difference of 1 degree can be decisive. If you begin to see how complex this whole arrangement is, you are beginning to gain an appreciation for a strong-performing port.
Another important element in all this angle interplay is the position and width of the intake seat on the valve. Positioning the intake seat as far to the outside diameter of the intake valve as possible improves airflow. Most head porters place this seat approximately 0.020 inch inboard of the outside diameter of the valve. To create a durable intake-sealing surface, most performance cylinder head builders specify an intake seat width of between 0.060 and 0.070 inch. A narrower seat width definitely improves airflow, but for a street engine, normal driving quickly pounds this out to a much wider seat that only results in lost flow.
A valve’s margin is defined as the thickness of the valve between the chamber side of the valve and the end of the valve sealing face. A margin of at least 0.050 inch is also a good idea and has been used to help create improved flow. Valve manufacturers often specify different margins for their valves since this additional material adds to the overall weight of the valve, and a couple of grams here and there add up pretty quickly. If there is additional flow to be gained, some head builders actually specify a slightly larger valve and then machine the outside diameter down by 0.020 inch or so in order to increase the margin thickness. This additional margin can also then be used to add a 20 to 25-degree top cut in an effort to improve flow. This is a less popular modification for intake valves, and we recommend testing this before blindly adding yet another angle to the flow mix.
Earlier in this discussion of intake flow, we mentioned that a majority of the flow exiting the intake port tends to push toward the roof side of the port and valve exit. If you look at the chamber arrangement of a typical 23-degree smallblock head mounted on a cylinder block, you’ll notice that the flow is predominantly aimed at the combustion chamber wall and then directly into the cylinder wall. This shrouding effect is especially bad on production iron heads with deep chambers. Experiments with top cuts on the combustion chamber can also be worth increased low-lift flow, but again there is less experimentation in this area to support exact details. We leave a more detailed description of chamber configuration to Chapter 11 where we get much deeper into this subject. But suffice it to say that the chamber side of the valve is a big-time player in not only combustion efficiency, but port flow numbers as well.
The exhaust side of the cylinder head always seems to attract less attention than the intake. Perhaps this is because the exhaust side is less “romantic” or maybe it’s just because the topic involves lots of hot air, but there’s no disputing the fact that, given the air pump analogy, you can’t get more good air in if you can’t get the bad air out. Dealing with exhaust flow is also substantially different from inlet air because we’re dealing with a high-temperature, high-pressure working fluid. Because of this, the exhaust valve can be smaller, roughly 80 percent of the intake valve diameter.
Ironically, on the flow bench, we flow test the exhaust port much like the intake except in the opposite direction. The temperature and the test pressure are usually the same as the intake. Perhaps because of these differences, the industry has created its own unique way of measuring the efficiency of the exhaust by comparing it to the intake side. This E/I (exhaust-to-intake) index is a percentage created by dividing the exhaust flow by the intake port flow at the same valve lift to come up with one type of rating system for the exhaust port.
One interesting point about this is that the exhaust rarely achieves more than around 80 to 85 percent flow compared to the intake port. Generally, if the exhaust port achieves this value, it’s not because the exhaust port is that great, but because the intake port flow value used for the comparison at that valve lift is generally weak. There is also much contention and discussion around what the “optimal” E/I value is for a particular cylinder head or engine family. Generally, these numbers revolve around the 75 to 80 percent value, but we don’t lay claim to any hard and fast rule. If the exhaust port is too efficient, it tends to overscavenge the cylinder and pull fresh inlet air right out of the port during overlap. If this occurs, the solution is generally to reduce cam overlap and/or reduce exhaust camshaft duration.
Exhaust port size is also critical to engine performance. Just as velocity is important to the intake side, it is an essential component on the exhaust side as well. Just like all other components for a street head, the port designer is forced to fit his exhaust port exit within a relatively narrow area of real estate in the cylinder head. Port exit location is dictated by the location of the exhaust header flange bolts in order for the headers to fit the chassis. This means the designer cannot arbitrarily raise the exhaust port location to improve exhaust flow such as could be accomplished in a pure race head. The width of the port is also limited because of the location and factory spacing of the exhaust bolts holes. But still, within this area there is much that can be done to improve exhaust flow. This is evident by the wide variety of flow numbers generated by the different cylinder head manufacturers. It appears that some cylinder head companies spend much more time on their exhaust ports than others. This is revealed not necessarily by the E/I number, but by the actual flow numbers themselves. What we are looking for is coefficient of flow numbers that take into account the cross-sectional area and the actual flow numbers.
If the cross-sectional area of the exhaust port is too large, velocity is reduced and it becomes more difficult for the exhaust port to purge all existing combustion residue from the cylinder. If the port cross-sectional area is too small, it’s easy to see how the port itself is a restriction, and more exhaust residue will remain in the cylinder. Either way, peak power is limited because at higher engine speeds, there is precious little time for the exhaust port to do its job. Cylinder displacement plays a part in this scenario as well since a larger cylinder with a longer stroke has the potential to ingest more air and fuel, consequently requiring a larger exhaust port to help evacuate the cylinder of this combustion residue.
It’s also important to think about the combustion process and what occurs along each step of the fourstroke cycle to truly appreciate the functionality required of each component on the cylinder head. In a performance engine, long before the piston reaches BDC, the exhaust valve opens to begin the exhaust phase. The earlier the exhaust valve opens, the more time it has (in terms of engine degrees of rotation) to help blow or push most of the exhaust gas out of the cylinder. This requires more exhaust lobe timing, which means opening the exhaust valve sooner and closing it later. A longer exhaust duration helps an under-performing exhaust port to improve top-end power, but this sooner opening exhaust valve point also potentially robs the engine of midrange power with the exhaust valve opening too soon at lower engine speeds.
Here’s where we get to discuss an interesting concept making the rounds in high-output naturally aspirated engine applications that value in terms of achieving the greatest reduction of exhaust gas volume in the cylinder. Many drag race cylinder head designers and builders ignore low-lift exhaust flow rates, claiming they are of insignificant value, and perhaps they know something about race engines that we don’t. Several cylinder head manufacturers do not even list 0.100-inch valve lift numbers in their catalogs, choosing to start the flow curves at 0.200 inch. However, there are a few respected cylinder head specialists who contend the low-lift exhaust flow numbers are important. In fact, one cylinder head porter we spoke to stated he’d prefer if we didn’t discuss this in our book since he was concerned about acknowledging this point to the benefit of other cylinder head companies! So perhaps there is something to this concept.
As far as street cylinder heads are concerned, even 0.100 inch of valve lift becomes critical if you consider that peak valve lift often occurs at 0.450 to 0.500 inch, making even 0.100 inch represent 20 percent or more of the total valve lift figure. Port work is often a game of compromises. It’s rare to find a change that is worth improved flow throughout the entire valve lift curve. More often it’s a case where a slight change in the seat angles or throat approach angle improve the flow at low lift while hurting high-lift flow. The opposite can also be true. Taking this low-lift flow concept one step further, it’s possible that sacrificing peak valve lift flow to enhance flow at 0.050 or 0.100 inch of valve lift might just be beneficial.
Further evidence supporting this low-lift flow theory states that incylinder pressure analysis of a race engine indicates that the majority of exhaust port flow occurs before the exhaust valve reaches peak valve lift. This also indicates that only a minor amount of flow occurs on the closing side of the exhaust valve curve. While this is only one test on one engine, it does indicate that this is a fairly typical situation that is probably duplicated in most every street engine.
One detracting component of improving low-lift exhaust flow is that most camshafts tend to accelerate the valve through 0.100 inch rather quickly, which is why certainly 0.200 inch valve lift is also an important exhaust flow number as well. Flat tappet cams are not limited in acceleration, so they pump the lifter through the 0.100-inch valve lift point rather quickly compared to roller cam followers that are a bit more lazy off the base circle. This would then mean that, in theory, good low-lift flow numbers would be of more positive consequence for roller cam motors than flat tappet cammed engines.
Another point worth considering is valve overlap, where both the intake and exhaust valves are open at the same time. This occurs as the exhaust valve is closing and the intake valve is just opening. It’s possible that exhaust gas exiting the cylinder past the exhaust valve tends to enhance intake low-lift flow. The negative side of this is that a strong low-lift flowing exhaust valve might also pull a portion of the inlet charge directly from the intake right out the exhaust. This probably does occur, but could be minimized by reducing the amount of valve overlap that is designed into the camshaft. As you can see, this is very much an openended discussion with proponents and detractors on both sides.
We only know what works for us on many different naturally aspirated small-block Chevys. While we have never actually tested the concept of improved low-lift exhaust port flow, we do know that every cylinder head that offers an excellent performing exhaust port in the low lift region tends to perform very well and usually requires less cam duration on the exhaust side than heads that do not offer good low-lift exhaust port flow numbers. To support this statement, we canvassed every single head in our flow inventory and came up with a list of five heads that offered no less than 67 exhaust cfm at 0.100-inch valve lift. A couple of heads, like the AFR 210 and the TFS 195 head, tested at 70- plus cfm at this same valve lift. All of these heads also perform in the top 10 percent of all the castings that we’ve tested. This example encompasses much more than just low-lift exhaust flow, but certainly this component must be a contributing factor.
Getting back to more specific app-lication recommendations, exhaust ports with a taller exit exhibit far superior flow characteristics than ports forced to pinch off the exit line for the exhaust gas. Some cylinder head manufacturers even raise the ports to the extent that headers no longer fit the chassis they were designed for. Another trick employed by the cylinder head porter to increase flow is to widen the exhaust port floor to even out the pressure distribution from the port into the header.
Because the exhaust port does not have to be concerned with wet flow characteristics, a full radius valve job can often be employed on the exhaust side with favorable results. These valve jobs enhance low and mid-lift flow by using smooth transitions between the valve angles. While this works on the exhaust side, don’t try this on the intake— the results will not be favorable. A wider margin on the exhaust valve is always preferable not just for the additional material as a heat sink, but also to enhance flow around the valve. Some head specialists prefer to fully radius the chamber side of the exhaust valve, while other ports respond negatively to this technique. A wider seat of around 0.100 inch is a good idea, especially because the valve seat along with the valvestem are the two primary heat transfer areas that pull heat out of the exhaust valve.
We discuss back cutting the valve in more detail in Chapter 10, but this is a quick way to enhance exhaust port flow. A 30-degree back-cut placed inboard of the 45-degree seat on the valve tends to especially enhance lowlift flow, and often it can be beneficial throughout the entire lift curve. If you have the time and resources to test, it may also be of value to try other back-cut angles besides 30 degrees. The guys at Lingenfelter Performance Engineering have found that even one-degree changes can be beneficial, but this requires precise and repeatable testing techniques to verify. But that’s how the fast guys become fast.
As we began this chapter, it became apparent that there was no way to adequately cover this subject in the amount of room we originally budgeted. In fact, this chapter is much larger than intended because we kept coming across more information that we couldn’t resist including. Our goal was to hit the major points of cylinder head airflow characteristics, but it is impossible to cover everything. There is clearly enough material for a complete book just on the physics and theory of airflow in an internal combustion engine. It was our intent to introduce you to this subject and perhaps open enough doors to intrigue you to continue your personal search for increased knowledge on the subject. If so, then this chapter has been successful. It was definitely fun to write.
Written by Chris Petris and Posted with Permission of CarTechBooks